Main Challenges of Price Volatility in Agricultural Commodity Markets

Chapter 2Main Challenges of Price Volatilityin Agricultural Commodity Markets

Monika Tothova

Abstract Prices of agricultural commodities undergoing rapid adjustments werein the spotlight following the “food crises” in late 2007 and early 2008, and againmore recently in summer and fall of 2010, raising concerns about increased pricevolatility, whether temporal or structural. Although price volatility is a normal fea-ture of markets given the seasonal production cycle and discontinuity of supply inthe face of a continuing demand, a greater uncertainty of a rapidly changing eco-nomic and natural environment contributes to and magnifies its occurrence. Thischapter focuses on the main challenges of price volatility in agricultural commoditymarkets. We start by briefly touching upon the theoretical aspects of volatility, fol-lowed by a comparison of international and European markets to identify whetherone was more affected than the other by increases in price volatility. Factors, impli-cations and preliminary policy considerations of increased volatility follow beforeinitial conclusions on future prospects are drawn.

2.1 A Primer on Theoretical Aspects of Volatility

Volatility provides a measure of the possible variation or movement in a particulareconomic variable. Prices change as rapid adjustments to market circumstances.Wide price movements over a short period of time typify the term “high volatility”.What constitutes a volatile market or an “excess volatility” can be subjective, sectorand commodity-specific.

Two measures of volatility are used:

Historical (realised) volatility, indicating a volatility of an asset in the past, isbased on observed (realised) movements of price over an historical period.It represents past price movements and reflects the resolution of supply anddemand factors.

Disclaimer: The views expressed in this chapter are those of the author and should not be attributedto her affiliated institution.

M. Tothova (B)Directorate General for Agriculture and Rural Development, European Commission,Brussels, Belgiume-mail: [email protected]

13I. Piot-Lepetit, R. M’Barek (eds.), Methods to Analyse Agricultural Commodity PriceVolatility, DOI 10.1007/978-1-4419-7634-5_2, C© Springer Science+Business Media, LLC 2011

14 M. Tothova

Implicit volatility,1 is the markets’ view on how volatile an asset will be inthe future. It represents the market’s expectation of how much the price ofa commodity is likely to move and tends to be more responsive to currentmarket conditions.

This chapter discusses historical volatility and does not refer to implicit volatil-ity. However, historical volatility can also serve as an indicator of the possibleprice changes of the assets in the future. Assets – including commodities – thathave high volatility are likely to undergo larger and more frequent price changesin the future, possibly attracting market participants benefiting from frequent pricechanges. A casual link between volatility and uncertainty is not clearly defined:volatility thrives in the environment of uncertainty, and volatile prices themselvescontribute to uncertainty for producers, processors and consumers.

A variety of measures is used to detect historical volatility, some of which arereferred to in the next section.

First Challenge of Volatility: Choice of Data for Analysis: Which Type,Frequency?

Choice of representative prices to analyse price volatility and their frequency is ofcrucial importance. Data with higher frequency exhibit higher volatility. Volatilitydecreases with decreasing frequency. Cash (spot) prices, such as CIF (cost, insur-ance, freight), can bring additional uncertainties to the analysis since transport pricesalone are very variable, influencing the result. FOB (free on board) prices are bettercandidates.

Commodity exchanges provide a steady stream of daily settlement prices makingthem ideal for analysis. However, futures markets do not exist or are not used for allcommodities. In addition, some contracts, such as the wheat contract on the ChicagoBoard of Trade, suffer from lack of convergence between cash and future prices.

2.2 Analysis: Is There More Volatility Now?

This section looks at spot and future prices using two different approaches todetermine whether the amount of volatility on agricultural commodity marketshas increased.2 Differences in price volatility on EU and world markets are alsodiscussed.

1This is calculated from the Black–Scholes formula for the price of a European call option on astock.2Volatility and variation are used interchangeably in this chapter.

2 Main Challenges of Price Volatility in Agricultural Commodity Markets 15

2.2.1 Spot Prices: An Intuitive Approach

For the “intuitive approach”-based deviations from a trend in prices, monthlyprice series from January 1997 to October 2010 on the EU and world agriculturalcommodity markets are analysed to determine whether:

1. World markets experienced higher price variation than EU markets, or2. Price variation on international and EU commodity markets increased over time.

EU data were taken from Agriview’s EU market prices3 for representative prod-ucts, and international commodity prices from international benchmarks from theWorld Bank or FAO (Food and Agriculture Organization). Some commodities mightnot be directly comparable in terms of quality and in some cases price data were notavailable on both world and the EU markets. Data and sources are described inAppendix 1. As discussed in the first challenge, monthly frequency can hide moreserious volatility issues by averaging daily data. In addition, international referenceprices for soybeans and soybean meal are CIF and thus include freight cost.

Two indicators are calculated:

1. A percentage of price observations lying outside the 20% tunnel around the pricetrend line. Using this method, observations just slightly over the trend line arecounted the same way as peaks.

2. A coefficient of variation as a ratio of standard deviation over mean as a measureof dispersion of data points. The higher the coefficient of variation, the larger thedispersion of series and the higher the price volatility.

Tables 2.1 and 2.2 summarise the results for relatively comparable products.Table 2.1 shows percentage of observations lying outside the 20% tunnel aroundthe trend line. Table 2.2 shows coefficient of variations.

Table 2.1 shows that over the studied period from January 1997 to October 2010,the percentage of observations outside the 20% tunnel was higher on the worldmarkets than on the EU markets (with the exception of chicken). However, differ-ences are noticeable in the cases of butter and Skim Milk Powder (SMP) where thepercentage of observations outside the 20% tunnel was significantly higher on theworld markets than in the EU (70+% compared to 20–30%). Dividing the data intotwo equally sized intervals (January 1997–November 2003 and December 2003–October 2010), we note that the percentage of observations outside the 20% tunnelon the world market exceeded the percentage of observations outside the 20% tun-nel on the EU market for barley, wheat, maize, butter, SMP and beef during the firsttime period. During the second time period, from December 2003 to October 2010which also included the periods of price hikes, we note that the absolute percentage

3http://www.agriview.com/

16 M. Tothova

Table 2.1 Twenty percent tunnel, comparable products

20% tunnel World prices EU prices

Commodity01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

Barley 66.87 55.42 78.31 57.93 28.40 86.75Wheat

(Int. SRW,EU bread)

72.89 75.90 69.88 54.88 28.40 80.72

Maize 62.05 67.47 56.63 48.78 19.75 77.11Butter 80.00 85.54 74.39 25.30 0 50.60SMP 72.29 79.52 65.06 28.92 16.87 40.96Chicken 10.84 13.25 8.43 15.66 9.64 21.69Beef 22.89 22.89 22.89 6.02 6.02 6.02

The bold figures indicate that “volatility” (measured either as CV or number of observations outsidethe 20% tunnel) increased in the second period

of observations outside the 20% tunnel on the world market decreased for all prod-ucts except for barley. On the EU market, the percentage of observations outside the20% tunnel increased for all commodities except beef.

Table 2.2 summarises coefficients of variations for the products discussed inTable 2.1. Comparing coefficients of variation on the world and EU markets cover-ing period from January 1997 to October 2010, we observe that prices on the worldmarkets were more dispersed than prices on the EU markets, with meats being lessdispersed than crops and dairy. On both the world and EU markets, the coefficient ofvariation increased between 1997–11/2003 and 12/2003–2010, indicating increaseddispersion of prices. However, with the exception of chicken, world markets expe-rienced more dispersed prices in the first period between 1997 and 2003 than EUmarkets did.

Table 2.2 Coefficient of variation, comparable products

Coefficientof variation World prices EU prices

Commodity01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

Barley 34.04 15.42 31.05 20.80 6.39 26.26Wheat

(Int. SRW,EU bread)

38.92 17.32 33.23 21.44 5.82 27.54

Maize 33.68 11.96 30.17 18.52 5.64 23.23Butter 46.56 16.93 35.72 10.55 3.47 12.84SMP 39.63 17.66 33.03 14.39 8.35 18.31Chicken 13.90 5.57 8.42 10.71 6.15 9.28Beef 21.26 9.77 13.00 6.84 4.07 5.40


Even in the second time period, with the exception of chicken, the coefficient ofvariation in the world price series exceeded the coefficient of variation in the EU.Price charts for comparable products are presented in Fig. 2.1.

Tables 2.3 and 2.4 show both number of observations outside the 20% tunnel andcoefficient of variation for world (Table 2.3) and EU (Table 2.4) prices for whichrespective equivalents were not identified. On the international markets, the per-centage of observations outside the 20% tunnel increased for sorghum, soybeanmeal and Whole Milk Powder (WMP). Coefficient of variation increased for allproducts between both sub-time periods, indicating higher dispersion of prices after2003. On the EU market, the percentage of price observations outside the 20% tun-nel more than doubled between both sub-time periods for crops and dairy, increasedsomewhat for eggs and decreased for most of the meats. The coefficient of varia-tion increased significantly for crops and cheeses while it decreased for meats andremained relatively stable for eggs.

Although the intuitive method suffers from a number of shortcomings (e.g. it failsto properly account for seasonality), it allows one to draw preliminary conclusions:

1. Using both the number of observations outside the 20% tunnel and the coeffi-cient of variation, from January 1997 to October 2010 world commodity marketsexperienced more volatility than EU markets. Coefficient of variation increasedboth on the world and EU markets between 01/1997–11/2003 and 12/2003–10/2010, with the EU recording more dramatic increases. However, in absoluteterms the coefficient of variation remains higher on the world markets than on theEU markets during 12/2003–10/2010 for all products except for chicken wherethe levels are comparable.

2. Compared to 01/1997–11/2003, dispersion of prices in 12/2003–10/2010 mea-sured by coefficient of variation increased for all commodities studied both in theEU and the world, with the exception of some meat products in the EU. However,compared to crops and dairy, volatility of meat prices is relatively low. Note thatthe latter time period includes price peaks, significantly shifting the mean of thetime series.

2.2.2 Volatility on the Futures Markets

Commodity exchanges produce a stream of daily settlement data. The use of nearbyfutures is also justified by frequently using those futures as international referenceprices. The Chicago Mercantile Exchange (CME) group offers already calculatedmeasures of volatility.4 For consistency for European exchanges we used settlementprices and the formula applied in the CME calculations for the milling wheat (from

4http://www.cmegroup.com/market-data/reports/historical-volatility.html. To annualize theirvolatility figures, the CME group uses an average of 252 trading days each year. Due to holidays

18 M. Tothova

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

Janu

ary 1

997

June

1997

Novem

ber 19

97

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 20

02

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 20

09

May

201

0

Octobe

r 201

0

US

D/t

Data Source: FAO

0

50

100

150

200

250

300

350

400

450

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 19

99

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 20

04

July

2004

Decem

ber 2

004

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 2

007

April 2

008

Septe

mbe

r 200

8

Febru

ary 20

09

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

eur/

100k

g

Data Source: AGRIVIEW

SMP

0

1000

2000

3000

4000

5000

6000

Janu

ary 1

997

June

199

7

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

2002

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

2007

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 20

09

May

201

0

Octobe

r 201

0

US

D/t

Data Source: FAO

0

50

100

150

200

250

300

350

400

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 19

99

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 20

04

July

2004

Decem

ber 2

004

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 2

007

April 2

008

Septe

mbe

r 200

8

Febru

ary 20

09

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

eur/

100k

g


Chicken

0

20

40

60

80

100

120

140

160

180

200

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 20

04

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 20

09

July

2009

Decem

ber 20

09

May

2010

Octobe

r 201

0

cen

ts/k

g

Data Source: WB

0

50

100

150

200

250

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 20

09

May

2010

Octobe

r 201

0

eur/

100k

g c

arca

ss w

eig

ht


Beef

0

50

100

150

200

250

300

350

400

450

Janu

ary 1

997

June

1997

Novem

ber 19

97

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 20

02

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 20

09

May

201

0

Octobe

r 201

0

cen

ts/k

g

Data Source: WB

0

50

100

150

200

250

300

350

400

Janu

ary 1

997

June

199

7

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 19

99

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

2002

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 2

004

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 20

07

June

2007

Novem

ber 2

007

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

eur/

100k

g c

arca

ss w

eig

ht


EUWorld Barley

0

50

100

150

200

250

300

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 19

99

May

200

0

Octobe

r 200

0

Mar

ch20

01

Augus

t 200

1

Janu

ary 2

002

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch20

06

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 20

09

May

201

0

Octobe

r 201

0

US

D/t

Data Source: WB

0

50

100

150

200

250

Janu

ary 1

997

June

1997

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch20

01

Augus

t 200

1

Janu

ary 20

02

June

2002

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 20

07

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

eur/

t


Wheat (Int. SRW, EU bread)

0

50

100

150

200

250

300

350

400

450

Janu

ary 19

97

June

199

7

Novem

ber 1

997

April 1

998

Septem

ber 1

998

Febr

uary 199

9

July

1999

Decem

ber 1

999

May

200

0

Octob

er 2

000

Mar

ch 200

1

Augus

t 200

1

Janu

ary 20

02

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 2

004

May

200

5

Octob

er 2

005

Mar

ch 200

6

Augus

t 200

6

Janu

ary 20

07

June

200

7

Novem

ber 2

007

April 2

008

Septe

mbe

r 200

8

Febr

uary

200

9

July

2009

Decem

ber 2

009

May

2010

Octob

er 2

010

US

D/t

Data Source: WB

0

50

100

150

200

250

300

Janu

ary 19

97

June

199

7

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 19

99

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

2002

Novem

ber 20

02

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

2005

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

200

7

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 20

09

July

2009

Decem

ber 20

09

May

201

0

Octobe

r 201

0

eur/

t


Maize

0

50

100

150

200

250

300

350

Janu

ary 1

997

June

199

7

Novem

ber 1

997

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 2

002

June

2002

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 2

004

May

200

5

Octobe

r 200

5

Mar

ch 2

006

Augus

t 200

6

Janu

ary 2

007

June

2007

Novem

ber 2

007

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

US

D/t

Data Source: WB

0

50

100

150

200

250

Janu

ary 1

997

June

199

7

Novem

ber 19

97

April 1

998

Septe

mbe

r 199

8

Febru

ary 1

999

July

1999

Decem

ber 1

999

May

200

0

Octobe

r 200

0

Mar

ch 2

001

Augus

t 200

1

Janu

ary 20

02

June

200

2

Novem

ber 2

002

April 2

003

Septe

mbe

r 200

3

Febru

ary 2

004

July

2004

Decem

ber 20

04

May

200

5

Octobe

r 200

5

Mar

ch20

06

Augus

t 200

6

Janu

ary 2

007

June

2007

Novem

ber 20

07

April 2

008

Septe

mbe

r 200

8

Febru

ary 2

009

July

2009

Decem

ber 2

009

May

201

0

Octobe

r 201

0

eur/

t


Butter

Fig. 2.1 Price charts with trend lines for comparable products


Table 2.3 World prices: 20% tunnel, coefficient of variation

20% tunnel 20% tunnel Coefficient of variation

Commodity01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

HRW wheat 65.66 71.08 60.24 37.75 16.06 31.91Rice Thai 5% 89.16 90.36 87.95 46.30 23.39 39.97Sorghum 59.04 54.22 63.86 31.19 11.60 28.33Soybeans 72.89 79.52 66.27 34.89 16.70 27.58Soybean oil 74.70 87.95 61.45 41.40 23.81 33.08Soybean meal 75.30 72.29 78.31 37.25 21.60 30.80Cheese 57.83 62.65 53.01 37.42 10.53 27.59WMP 73.49 69.88 77.11 40.15 13.12 33.91

Table 2.4 EU prices: 20% tunnel, coefficient of variation

20% tunnel Coefficient of variation

Commodity01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

01/97–10/10(%)

01/97–11/03(%)

12/03–10/10(%)

Feed wheat 56.10 28.40 83.13 23.18 7.37 29.47Durum wheat 59.15 39.51 78.31 33.81 12.08 39.59Malting barley 59.15 32.10 85.54 23.51 6.02 27.79Cheddar 50.00 34.94 65.06 17.38 5.59 16.63Eidam 27.11 10.84 43.37 8.96 4.60 10.21Young

bovines9.64 18.07 1.20 8.98 7.07 6.48

Cows 24.70 39.76 9.64 9.50 7.77 6.35Heifers 1.81 3.61 0 7.47 4.66 5.31Piglets 59.64 77.11 42.17 19.28 24.49 12.07Pork 31.93 50.60 13.25 13.15 17.02 7.80Eggs 43.37 36.14 50.60 16.18 13.71 16.59

September 1998 to October 2010) contract on MATIF. The formula is outlined inAppendix 2.

Different products (wheat, maize, oats, soybeans and derived products) show dif-ferent price and volatility patterns. However, there are commonalities across them.Although increased volatility can occur in any given period, actual peaks differ onthe basis of the commodity and developments of their fundamentals. Due to spacelimitation we would focus on wheat in this chapter. Figure 2.2 shows historicalvolatility of wheat on the Chicago Board of Trade (CBOT – part of the CME group)

and weekends, the number of actual trading days each year can differ, and as such volatility resultscan differ.

20 M. Tothova

0

10

20

30

40

50

60

70

80

Janu

ary 1

980

Novem

ber 1

980

Septe

mbe

r 198

1

July

1982

May

198

3

Mar

ch 1

984

Janu

ary 1

985

Novem

ber 1

985

Septe

mbe

r 198

6

July

1987

May

198

8

Mar

ch 1

989

Janu

ary 1

990

Novem

ber 1

990

Septe

mbe

r 199

1

July

1992

May

199

3

Mar

ch 1

994

Janu

ary 1

995

Novem

ber 1

995

Septe

mbe

r 199

6

July

1997

May

199

8

Mar

ch 1

999

Janu

ary 2

000

Novem

ber 2

000

Septe

mbe

r 200

1

July

2002

May

200

3

Mar

ch 2

004

Janu

ary 2

005

Novem

ber 2

005

Septe

mbe

r 200

6

July

2007

May

200

8

Mar

ch 2

009

Janu

ary 2

010

Per

cen

t

Data source: CME Group

Fig. 2.2 US wheat, historical volatility, monthly annualised

on a monthly basis from January 1980 to October 2010. Wheat volatility has had anincreasing trend over the observed period, ranging between 30 and 73%. In the last4 years the average volatility has increased.

MATIF wheat experienced the highest volatility in September 2007, January2009 and July–August 2010 when it reached around 44–48%. The summer2010 high volatility episode accompanied poor harvest prospects in Russia andconsequent export ban. However, in between those peaks, the volatility was as lowas 8% (February 2010). Although experiencing peaks, wheat volatility on MATIFwas relatively stable between 1998 and mid-2006 when it started increasing.

Second Challenge of Volatility: Choice of Method and Reference Period

A variety of approaches to detect volatility yielding different results are in use.Different results are also obtained using different reference periods for comparison.Crude methods applied in this chapter showed that price volatility is increasing.However, even though the presence of volatility was not increasing over a longertime frame, it is important to compare shorter time frames.

Although this chapter has not looked at the long-term data series, others (e.g.OECD/FAO, 2010) have done so and did not support a case for decreasing long-term volatility trend as the current boom of volatility does not match the heightsreached in the 1970s. While correct on technical grounds, the findings are not ofimmediate relevance to producers who were faced with lower price variability inthe preceding two decades.


Wheat: MATIF Historical Volatilitymonthly annualised

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

Septe

mbe

r 199

8

Janu

ary1

999

May

199

9

Septe

mbe

r199

9

Janu

ary 2

000

May

200

0

Septe

mbe

r 200

0

Janu

ary 2

001

May

200

1

Septe

mbe

r200

1

Janu

ary 2

002

May

200

2

Septe

mbe

r 200

2

Janu

ary2

003

May

200

3

Septe

mbe

r 200

3

Janu

ary 2

004

May

200

4

Septe

mbe

r 200

4

Janu

ary2

005

May

200

5

Septe

mbe

r 200

5

Janu

ary 2

006

May

200

6

Septe

mbe

r 200

6

Janu

ary 2

007

May

200

7

Septe

mbe

r 200

7

Janu

ary 2

008

May

200

8

Septe

mbe

r 200

8

Janu

ary 2

009

May

200

9

Septe

mbe

r 200

9

Janu

ary 2

010

May

201

0

Septe

mbe

r 201

0

Per

cen

t

Data source: MILLING WHEAT #2: 1ST EXPIRATION FUTURE NEARBY - SETL - MARCHE A TERME INTERNATIONALE DE FRANCE (MATIF) via Global Insight. In-house calculations.

Fig. 2.3 MATIF milling wheat, historical volatility, monthly annualised

Agricultural commodities traded on European exchanges, although smaller interms of volume, were not shielded from increased volatility. Figure 2.3 shows thedevelopment of historical volatility for milling wheat on MATIF.

2.3 Factors Influencing Price Volatility

Price volatility is driven by the same set of factors as prices – a topic studied in detailduring and following the price hikes of 2007–2008 (e.g. EC, 2008; Meyers, 2009;Trostle, 2008). Among the wide variety of factors are underlying market fundamen-tals such as yields and stock levels; weather and changing weather patterns withtheir related impacts; cycles in key markets; policy driven developments includinglarge purchases by the governments; developments outside the agricultural sectorsuch as exchange rate and oil price movements; trade policies and their transmis-sion; investment in agricultural production etc. Commodities for which the demandis inelastic (such as agricultural products) tend to be more volatile. Long-term struc-tural changes are also responsible for the increase in price variability, although theireffects are not immediate. Only some of the factors contributing to greater volatilityare described below.

Low levels of stocks in their own right do not result in high price but providea limited buffering capacity should increasing demand or short-term supply chal-lenges occur. There is no single answer to the question “What normal stocks are?”.

22 M. Tothova

In addition, stock management, such as stock creation and release, can affect marketfundamentals and impact prices.

Climate change and weather-related events impact production variability andthus impact market fundamentals. So far on the EU level, no correlation has beenestablished between the warming of the last decades and the level of crop yields,which have generally increased (EC, 2009). However, the impact of climate changemight be visible already in other, more vulnerable countries.

A frequent culprit of increased price volatility is speculation based on invest-ing in futures contracts on commodity markets to profit from price fluctuations.The wider and more unpredictable price changes are, the greater the possibility ofrealising large gains by speculating on future price movements of the commodityin question. Although the presence of “speculators” on the derivatives markets isa necessary condition for functioning markets and efficient hedging, volatility canattract significant speculative activity and destabilise markets, which are both thecause and effect of increased volatility. In thinly traded markets where only smallquantities of physical goods are traded, the value of speculative trades may cre-ate false trends and drive up prices for consumers. Arguments both for (e.g. Irwinand Sanders, 2010) and against (e.g. Robles et al., 2009) “speculation” are ample,although evidence is inconclusive. While other factors and fundamentals are at playand have to be considered, there is a time overlap between increased volatility andincrease in open interests on the commodity markets (Fig. 2.4). While increase inopen interests and inflow of investment money increases the liquidity on the market,increased liquidity could come with increased volatility.

0

500

1000

1500

2000

2500

3000

3500

4000

0

10

20

30

40

50

60

70

80O

pen

inte

rest

an

d v

olu

me

(th

ou

san

ds

of

con

trac

ts)

Vo

lati

lity

(%)

Futures, open interest Futures, volume Wheat volatility

Data source: CME Group

Fig. 2.4 US wheat historical volatility, futures open interest and volume monthly


Policies: Greater market orientation of agricultural policies (CAP included)relies on a greater transmission of market signals, and results in more variableprices. Policy instruments (described later) are in place to mitigate effects of pricevariability. Trade restrictive policies also play a role in limiting supplies, thusincreasing uncertainty on the markets and price variability.

Strong co-movements with energy and other agricultural prices: Linkages withenergy markets before the emergence of biofuels were one-way: oil and energy asinputs to agricultural production. Increased connection between energy and agricul-ture raises questions about volatility transmission from more volatile energy and oilmarkets in addition to changing market fundamentals, or at times without a signifi-cant change in market fundamentals. The strength of the link is not yet determined,although Du et al. (2009) found evidence of volatility spillover among crude oil,maize and wheat markets after autumn 2006 and explained it by tightened interde-pendence between these markets induced by ethanol production. Figures 2.5 and2.6 show scatter charts of daily settlement data for maize and crude oil for the2000–2004 and 2005–2010 periods. An OLS line fitted to the data reveals a strongercorrelation in the 2005–2010 time period with an R-squared of over 53% when notincluding a trend variable, and 56% when including a trend variable to avoid spu-rious regression, with all estimates significant at 5% level of significance. Scattercharts for data before 2000 (not included) resemble that of 2000–2004, with nosignificant correlation.

0

50

100

150

200

250

300

350

0.00 10.00 20.00 30.00 40.00 50.00 60.00

Oil Settlement

Mai

ze s

ettl

emen

t

Fig. 2.5 Scatter chart of maize and crude oil settlement prices, 2000–2004

24 M. Tothova

0

100

200

300

400

500

600

700

800

0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00160.00Oil Settlement

Mai

ze s

ettl

emen

t

Fig. 2.6 Scatter chart of maize and crude oil settlement prices, 2005–2010

2.4 Implications of Increased Volatility

Although references are often made to “excess volatility”, it is generally acceptedthat a certain degree of volatility is desirable, and price volatility is a normal featureof the markets. Without price adjustments, markets would come to a stall. Volatilityacross the commodity markets is not consistent. Although active participants onagricultural commodity markets are finding prices to be volatile, compared to energymarkets, volatility remains rather low. Energy returns have been significantly morevolatile than other commodity sectors. Other markets, such as metals, have expe-rienced higher volatility than energy markets; these episodes have been brief andtransitory.

In macroeconomic terms, while price hikes are beneficial for net exporting coun-tries that benefit from improved balance of payments, they increase the import billof net importing countries. Food security considerations play an important role.Variable prices lead to an uncertain food import bill, and high prices impact theability of poor consumers to purchase necessary food. On the other hand, producersand net sellers benefit from increased prices.

Concerns about increased price volatility are usually voiced by producers andprocessors who in the absence of risk management tools are exposed to unpre-dictability and uncertainty associated with changing prices. High fluctuations inprices may limit the ability of consumers (processors) to secure supplies and controlinput costs. Due to price transmission issues, contracting and relatively low percent-age of raw commodity in the processed products, consumer prices do not necessarilyfollow commodity prices directly. While we focus on the volatility of output prices,


volatility of input prices (oil, fertiliser etc.) also affects agricultural production anddecision-making.

The biggest drawback of volatility is the associated uncertainty of marketingproduction, investment in technology, innovation etc. The persistence in volatilityreflects the continued uncertainty with regards to how market fundamentals haveunfolded and how they are likely to unfold. Higher price volatility means highercosts of managing risks (such as higher margins on futures contracts and higherpremiums for crop revenue insurance). It is likely that higher costs of risk mitiga-tion would eventually translate into higher consumer prices. Commodity shocks inthe form of increased prices and increased volatility can also have an impact oninflation, although this chapter abstains from analyzing the link.

A distinction has to be made between the effects of volatility itself (such as unsta-ble prices and their impact on food security) and effects of policy reactions. Short-term policy reaction can contribute to market instability and consequently volatility,as we observed in the case of rice in spring 2008 when in the wake of increasingprice levels some major exporting countries introduced export restrictions, and againin summer 2010 following an export ban in Russia.

2.5 Policy Considerations

Increased volatility can be addressed in two different ways:

1. Dealing with price volatility itself by trying to stabilise markets using pricecontrols and supply controls (stock management).

2. Dealing with the effects of increased price volatility by employing risk manage-ment tools (crop insurance, [functioning] futures markets), income stabilisationmechanisms, safety nets etc.

2.5.1 Dealing with Price Volatility Itself

Price controls and supply controls go hand in hand. Past attempts to stabilisecommodity prices – and thus reduce volatility – using international commodityagreements, marketing boards, supply controls, planned economies, or more explicitprice setting nationally were not a great success. Since successful manipulation ofmarket fundamentals is unlikely, a safe – but rather slippery – way to reduce oreven eliminate volatility is to fix prices. However, such experiments in the past ledto various forms of market failures, leading to inefficiencies, and are unlikely togain broad support. Dawe (2009) discusses both costs and benefits of stabilisingprices of staple foods. The main cost of price stabilisation is the deadweight lossby not allowing market prices to follow world prices. Among the beneficiaries ofprice stabilisation are consumers and producers benefitting from stable prices. Inthis case price stabilisation serves as a safety net program, although not the mostefficient one.

26 M. Tothova

The international commodity agreements regulating supply through productionquotas for its members and maintaining buffer stocks, so that world prices remainstable, within a specified range were used from the 1950s to the 1990s. These agree-ments managed to sustain world prices for a number of products (notably coffee),but the eventual collapse brought about by competitive pressure from producercountries and a withdrawal of support from consumer countries has made themlargely ineffective at keeping prices level. For a detail on international commodityagreements, refer to Gilbert (1996).

Stock management played an important role in international commodity agree-ments. Following the price hikes of 2007–2008, many advocate the role of buildingstocks as a way to buffer against sudden changes in prices. Proposals currentlyon the table address increased volatility in an ad hoc fashion mostly addressingissue of stocks, both virtual and real. While stocks fulfill a buffering role, they alsoremove commodity from the market and thus at the times of tight supply mightput additional strain on it. Management of stocks also comes with a high cost ofgovernance.

It might be possible to deal with price volatility itself in the longer termby strengthening market fundamentals, by securing supply and by introducinginnovation and new technology.

2.5.2 Dealing with the Effects of Increased Price Volatility

Price volatility affects both macro and microeconomic aspects. On the macroeco-nomic level, price volatility influences balance of payments of both importing andexporting countries. If volatility attains a significant level, it may affect the abil-ity of governments to plan and provide economic security and economic growth.Although price hikes draw attention to net food importing countries that see theirimport bills soaring, the effects of price decreases are naturally felt in net exportingcountries.

Dealing with effects of increased price volatility calls for income stabilisation.One way to income stabilisation uses price stabilisation described earlier. However,price stabilisation is not a necessary condition for income stabilisation, which canbe achieved by designing efficient safety nets. In the developed countries with well-established agricultural policies, many programs already contain instruments to aidincome stabilisation. Where this is the case, it is important that income support bedecoupled from production to minimise production and trade distortions.

Commodity price risk management uses financial instruments for managingprice risks rather than reducing price volatility itself. Risks are not transferred tothe government but are reallocated among private traders. Among those instru-ments are futures and forwards contracts, commodity swaps, call and put options,commodity-indexed bonds and long-term contracts. There is renewed interest in therange of options based on market-based risk management instruments that mighthelp countries and individuals generate more stable and predictable incomes. Use


of market-based risk management instruments requires the proper functioning ofderivatives markets. However, while crop insurance and futures markets work rel-atively well in developed countries, extending them to developing countries mightnot always be feasible.

The backbone for successfully coping with commodity price volatility relieson strong institutions and management. A further development of market-basedinstruments that react to market signals, while at the same time helping to mit-igate the effects on incomes, is important. A possible development of financialderivates could also play a role. For this to take place, a transparent regulationof commodity exchanges is a pre-requisite. Safety nets could be developed, orwhere they exist, be reinforced to mitigate effects of volatile prices. Currently, manydeveloping countries are lacking safety nets as well as access to efficient savinginstruments. The best long-term solution to commodity price volatility would beproduct diversification.

Third Challenge of Volatility: When Is Volatility Excessive? What PoliciesShould Be Employed?

The level of volatility is commodity-specific, differing across sectors and commodi-ties within a sector. A question to answer is whether volatility should be prevented,risking obstruction of market signals, or whether addressing consequences of pricevolatility aiming at income stabilisation is more desirable.

2.6 Concluding Thoughts

Although volatility has always been a feature of agricultural commodity markets,the evidence suggests that volatility has increased at least in some commodity mar-kets. There seems to be an overlap between periods of high prices and increasedvolatility. Volatility peaks also seem to coexist with decreased stocks. The chapterabstained from considering the development of fundamentals and macroeconomicfactors, such as exchange rate developments.

Persistence of volatility points to uncertainty in developments of market funda-mentals coupled with structural and monetary policy. Higher price volatility meanshigher costs of managing risks (such as higher margins on futures contracts andhigher premiums for crop revenue insurance). However, with increasing biofuelsproduction, a tightened interdependence between crude oil and commodity mar-kets can be expected which could result in increased transmission of crude oil pricevolatility into agricultural commodity markets. It is likely that higher costs of riskmitigation would eventually translate into higher consumer prices.

Increased volatility highlights the presence of greater uncertainty on the mar-ket. Two broad sets of policies could be employed: (1) those that target volatilityitself, such as price and supply controls, and (2) those that deal with the effects ofprice volatility while letting markets work, such as risk management instruments,safety nets etc. Policies based on price and supply controls do not appear to havean impressive precedent, and reduce market signals. Policies mitigating volatility

28 M. Tothova

or the effects of volatility should aim to address uncertainties and focus on riskmanagement while keeping markets working.

It remains impossible to capture future price variability. However, elements fromthe past that resulted in the past and present variability remain active.

Appendix 1: Description of Price Series Used in Section 2.2

World grains, oilseeds, and meats: compilation of various sources by WorldBank Commodity Price Data (Pink Sheet), available at http://go.worldbank.org/2O4NGVQC00

– Barley (Canada), feed, Western No. 1, Winnipeg Commodity Exchange, spot,wholesale farmers’ price

– Wheat (US), no. 2, soft red winter, export price delivered at the US Gulf port forprompt or 30 days shipment

– Maize (US), no. 2, yellow, f.o.b. US Gulf ports– Wheat (US), no. 1, hard red winter, ordinary protein, export price delivered at the

US Gulf port for prompt or 30 days shipment– Rice (Thailand), 5% broken, white rice (WR), milled, indicative price based on

weekly surveys of export transactions, government standard, f.o.b. Bangkok– Sorghum (US), no. 2 milo yellow, f.o.b. Gulf ports– Soybeans (US), c.i.f. Rotterdam– Soybean oil (Any origin), crude, f.o.b. ex-mill Netherlands– Soybean meal (any origin), Argentine 45/46% extraction, c.i.f. Rotterdam begin-

ning 1990; previously US 44%– Meat, beef (Australia/New Zealand), chucks and cow forequarters, frozen bone-

less, 85% chemical lean, c.i.f. U.S. port (East Coast), ex-dock, beginningNovember 2002; previously cow forequarters

– Meat, chicken (US), broiler/fryer, whole birds, 2-1/2 to 3 pounds, USDA grade“A”, ice-packed, Georgia Dock preliminary weighted average, wholesale

World dairy prices: FAO compilation of average of mid-point of priceranges reported bi-weekly by Dairy Market News (USDA). Available athttp://www.fao.org/es/esc/prices/PricesServlet.jsp?lang=en

– Butter, Oceania, indicative export prices, f.o.b.– Cheddar Cheese, Oceania, indicative export prices, f.o.b.– Skim Milk Powder, Oceania, indicative export prices, f.o.b.– Whole Milk Powder, Oceania, indicative export prices, f.o.b.

EU market prices for representative products (monthly) Available at http://ec.europa.eu/agriculture/markets/


Appendix 2: Theoretical Consideration

The CME calculation of historical volatility calculation is the annualised standarddeviation of the first difference in the logarithmic values of nearby futures settlementprices. Mathematically, it can be written as

Volatility = STDEVDayNDay1

(ln

Settle PxT

Settle PxT − 1

)√252

where 252 is the estimated number of trade days in a year to convert volatility intoannualised terms.

References

Dawe D. (2009). Price stabilisation for staple foods: costs and benefits, implications for modelling,and using trade policy as a safety net. Paper prepared for a meeting on Uncertainty and pricevolatility of agricultural commodities, MOMAGRI, Paris.

Du X., Yu C.L., Hayes D.J. (2009). Speculation and volatility spillover in the crude oil and agricul-tural commodity markets: a Bayesian analysis. http://www.card.iastate.edu/publications/DBS/PDFFiles/09wp491.pdf

European Commission, EC (2008). High prices on agricultural commodity markets: situationand prospects. http://ec.europa.eu/agriculture/analysis/tradepol/commodityprices/high_prices_en.pdf

European Commission, EC (2009). Commission staff working document: adapting to climatechange: the challenge for European agriculture and rural areas. SEC (2009) 147. http://ec.europa.eu/agriculture/climate_change/workdoc2009_en.pdf

Gilbert C.L. (1996). International commodity agreements: an obituary notice. World Development,24, 1–19.

Irwin S.H., Sanders D.R. (2010). The impact of index and swap funds on commodity futures mar-kets: preliminary results. OECD Food, Agriculture and Fisheries Working Papers, 27, OECDPublishing. doi: 10.1787/5kmd40wl1t5f-en OECD.

Meyers W.H., Meyer S. (2009). Causes and implications of the food price surge. http://www.fapri.missouri.edu/outreach/publications/2008/FAPRI_MU_Report_12_08.pdf

OECD-FAO (2010). Agricultural outlook. Paris, OECD.Robles M., Torero M., von Braun J. (2009). When speculation matters. IFPRI. http://www.ifpri.

org/sites/default/files/publications/ib57.pdfTrostle R. (2008). Global agricultural supply and demand: factors contributing to the recent

increase in food commodity prices. http://www.ers.usda.gov/Publications/WRS0801/

http://www.springer.com/978-1-4419-7633-8

Main Challenges of Price Volatility in Agricultural Commodity Markets

Documents

Transcript of Main Challenges of Price Volatility in Agricultural Commodity Markets